Dr. Amy Kind is a geriatrician whose goal is to become an independent investigator in interventional health services research (i.e. clinical effectiveness research), focusing on transitional care safety in vulnerable older adult populations. The proposed award will facilitate this goal by expanding her competencies in clinical trials, statistical analysis, leadership, and systems engineering for health system re-design. Dr. Kind will access a wealth of resources through the University of Wisconsin, including accomplished NIH-funded mentors in the multi-disciplinary fields of clinical trials, systems engineering and health services research. The goal of Dr. Kind's proposed research is to address the $15 billion/year bounce-back problem while simultaneously improving the transitional care of sub-acute care patients with and without dementia. Bounce-backs are movements to settings of higher care intensity within 30 days of hospital discharge (i.e. rehospitalizations, emergency room visits). Existing effective interventions decrease bounce-backs through improved post-hospital communication, but use direct patient education to do so, which makes them inappropriate for most high-risk sub-acute care (largely hip fracture and stroke) and dementia populations. A systems approach which improves communication by optimizing the common hospital discharge summary has potential to reduce bounce-backs, but the most critical discharge summary components must be identified to enable intervention design. In this study we will utilize linked discharge summary-Medicare outcomes data to (1) examine the impact that omissions of specific discharge summary components have on post-hospital patient outcomes (bounce-backs, survival and cost) and identify the mechanisms by which these effects occur; and (2) determine how comorbid dementia modifies these impacts. Additionally, we will utilize a non-randomized prospective intervention study with historical and contemporary comparison groups to (3) determine the effect of a systems-engineering-based intervention vs. typical clinical practice on discharge communication rates of targeted components and on bounce-backs in hip fracture and stroke patients with and without comorbid dementia discharged to sub-acute care. PUBLIC HEALTH RELEVANCE: The proposed multi-disciplinary study is the first to address the common and costly problem of bounce-backs in vulnerable sub-acute care and dementia populations, and does so by using the common discharge summary; an innovative, cost-effective approach. This research will have important implications for transitional care interventions, development of quality measures and creation of patient care guidelines, and will provide essential information to heath systems, accreditation authorities and clinicians alike.